基于改进蚁群算法的个性化学习路径推荐

目前已有的学习路径推荐领域多为学习资源推荐,而课程知识图谱应用率较低,与蚁群算法的结合普遍缺乏对学习者知识水平的精确建模.因此,提出将知识图谱技术、深度知识追踪模型以及蚁群算法三者相结合,同时分类蚁群改进传统的蚁群算法:首先,抽象出课程知识点图谱作为路径基础,将深度知识追踪应用于不同水平学习者的分类,并与知识点难度权重相结合;然后,采用蚁群算法进行相应的路径规划,将蚁群按照不同的学习者类别进行划分,在保障相对最短学习路径的同时考虑不同学习群体客观知识水平情况,从而得到个性化的高效率学习路径推荐;最后,在ASSISTment数据集上验证了本方法的有效性.

Personalized learning path recommendation based on improved ant colony algorithm

At present, most learning path recommendation fields are learning resource recommendation and the application rate of course knowledge graph at a low rate. Therefore, a method which combines knowledge mapping technology, deep knowledge tracking model, and ant colonies, is proposed, to improve the classification of ant colonies in the traditional ant colony algorithm. Initially, taking a course knowledge map as a foundation, deep knowledge tracking is applied to the classification of different levels of learners and combined with knowledge difficulty weights. The corresponding path planning with ant colony algorithm, classifies ants according to different learner categories. The shortest path in considering objective knowledge groups of different learning levels is also studied to make personalized efficient learning path recommendation. Finally, the validity of the proposed method is verified on the open dataset of ASSISTment.